As a conventional technique for providing a communication service customized for each user, there has been disclosed a method for reducing the load of a service control program in a service control unit caused by transferring the information between service control units, by reducing an amount of individual user information to be maintained in the service control program. (For example, refer to the patent document 1.)
Also, there has been disclosed a method for providing a service customized for each user based on a processing policy provided in a service control unit which stores each user's condition and the processing policy set for individual users having the opposite user requesting for communication with a user, a condition of the user requested, and a process corresponding to a request content. (For example, refer to the patent document 2.)
Meanwhile, in recent years, there has been proposed a concept of Policy-Based Networking (PBN), which is a framework for controlling an IP network. In the PBN, a policy server sets network operation policies into a network apparatuses. By referring to the policies, the network apparatuses perform network services so as to meet QoS (Quality of Services) requirement, etc.
However, in view of setting a policy in each mobile terminal (user), it is required to set the policy to the entire apparatuses having possibilities of accommodating such a mobile terminal, which results in an increased amount of policy setting processing throughout the network. Further, in order to apply the information notified in the PBN to the individual basic services specified by the Mobile IP, etc., it is necessary to make a concrete specification to be applied to each service, as well as studies for implementation.
In order to avoid the aforementioned increase in the amount of policy setting processing, there may be considered a method by the use of a connection authentication procedure or a location registration procedure in a mobile protocol (for example, Mobile IP) performed by a user host terminal against the network. According to such a method, service control information for each user is included in a message transferred between the apparatuses having the host authentication procedure. This service control information is distributed to an edge router (router positioned in the edge zone of a core network). The edge router refers to the acquired service control information, and controls the service behavior based on the acquired service control information.
However, such the service control performed by the edge router is suitable for a service closed within the network layer (the layer three in the OSI reference model, or the IP layer).
As compared to the service performed in the above-mentioned layer, a service performed in the layer higher than the layer four, such as the layer seven (or the application layer), has features described below. (Hereinafter the service is referred to as ‘high-layer services’.)
Generally, the high-layer services is not dependent on a packet transfer path, etc. Such a service is not always appropriate to be performed in the edge zone of the core network.
Also, in general, it is not possible to identify whether or not the high-layer services is requested at the time of authenticating an access from a user terminal. For example, as for a user who started to use in a public wireless LAN service area, it is not possible to identify whether the user will use the IP telephone service first, or instead, access the Web service.
[Patent Document 1]
The official gazette of Japanese Unexamined Patent Publication Number Hei-8-256367 (pages 3-5, and FIG. 1)
[Patent Document 2]
The PCT Gazette of International Publication Number 00/19326